Last January, Jacquelyn Gill tweeted that she wanted to read a paper a day in 2015. I thought this was a great goal, and we decided we should create a hashtag to report on our reading efforts. Hence, #365papers was born. I didn’t really expect I would read a paper a day – and, at first, tweeted papers under both the #365papers and #260papers hashtags (where 260 had been proposed as a more reasonable number of working days in a year). But I (overly) optimistically thought there would be plenty of days where I’d read more than one paper, and that doing so would offset the days where I didn’t read any papers. I was completely wrong about that!* That said, I still think it was great to have the goal, as I think it got me reading more often, especially in the first half of the year.
Along the way, I kept track of what I read. Looking back through it has been interesting (to me, at least). Here’s more on what I read.
How did I do? Not as well as I’d hoped, but not completely terribly, in my opinion. Overall, I read 181 “papers” – though what to count was not always clear. I counted only papers that I read thoroughly and completely – say, at the level that I read something for a lab meeting. This meant that a lot of things that I read didn’t get counted, because I didn’t read the whole thing or only skimmed parts of it. I decided to count manuscripts and grant proposals that I was reviewing, as well as individual chapters of books and dissertations. In the end, I ended up reading:
- book chapters: 25
- journal articles: 100
- manuscripts or proposals: 49
- dissertation chapters: 7
The thing that stands out the most to me is how much of my (thorough) reading was for reviewing purposes. Some of that was as an associate editor, some as a regular peer reviewer, some as an ad hoc reviewer, and some as a panelist**. And, for papers I handled as AE, I counted each submission separately, because I read each submission really thoroughly (even if the paper just needed minor revisions.) So, the count would have been lower if I just counted each manuscript and not each submission. Still, 49 was higher than I expected, even given that.
Something else that surprised me was looking at the journals where the 100 journal articles that I read had been published. The only journals where I read at least 5 papers were Evolution, Nature, PNAS, Proceedings B, and Science. I definitely did not expect PNAS to be higher than Ecology or AmNat! The other surprising thing was how many journals there were: the 100 papers were in 53 different journals. I wouldn’t have expected that many, and suspect a lot of it relates to no longer primarily finding papers via emailed TOCs. Instead, I find most papers via Google Scholar alerts and, to a lesser extent, twitter.
If I’d kept track of the papers I skimmed or read only partially, the totals for AmNat, Ecology, and Ecology Letters would have been much higher. There are always at least one or two in each TOC that I receive that I skim, but apparently I don’t read those articles as thoroughly as I would have guessed. I’m not sure if that’s a problem or not.
Related to skimming: I realized, because of this project, how many articles I skim for teaching-related purposes. I read a lot of different things to get examples to use in class, but rarely read them thoroughly enough that I thought they should count. I guess that’s not surprising, but it really stood out to me, especially in the summer and fall.
I didn’t track gender of first authors during the year, but went back in and added that in after a discussion with Jacquelyn Gill and Anne Jefferson. Looking just at the journal articles, 39 had female first authors and 59 had male first authors. I had a sense that some of the male skew came from reading some older papers, so also looked just at papers published in 2010 or later. Of those, 37 had female first authors and 38 had male first authors. I’m not really sure what to make of that, but am happy that the new papers are well-balanced.
Year of publication
I was interested in seeing when the journal articles I read were published. I knew there were some really old ones, and suspected that overall it would skew heavily towards new papers. That was a pretty accurate assessment:
Figure 1. Year of publication of the journal articles I read in 2015, binned into 5 year intervals. 2015 was by far the most common year of publication.
Date of reading
The other thing that I was interested in – and suspected I’d be embarrassed by – was when I read things over the year. For this, I kept track of everything I read, including journal articles, manuscripts, proposals, and chapters. As I expected, I got off to a pretty strong start and then just completely fizzled in the fall.
Figure 2. Cumulative frequency distribution showing number of “papers” read over the course of the year. It definitely tapered off a lot in the fall.
I blame field season, teaching, and being on a faculty search committee for the tail off, but wish I’d done a better job of keeping up reading then. That said, there are only so many hours in a day (and only so many I can/want to devote to work).
Overall, my take is that this was an interesting experiment, and I plan to keep track of papers in 2016, too. However, I don’t think I’ll have 365 as my goal. When I got really off track, it became demoralizing to realize just how far off track I was. Instead, #200papers (suggested by my UMich colleague Tim McKay) seems like a more reasonable goal. Though, given the success of the #365papers hashtag, I might continue using that, even if 200 papers is actually my goal I like Abby Lawson’s idea to keep track of why I’m reading a paper (e.g., proposal, teaching, lab meeting, etc.) I also like Kirsty MacLeod’s idea to track how I found a paper (reference, twitter, TOC, etc). We’ll see how I do!
* I also set a goal to write a blog post a week. I reached my #52posts goal, so at least I was successful with one of them!
** I was on a preproposal panel, so counted 3 preproposals as a single “paper”, since they’re shorter.